Automatic detection of human's falls from heights for airbag deployment via inertial measurements

Abstract Workers in different industrial sectors, mainly in the construction and energy areas, face the daily risky situation of falling from the height accidentally, even wearing safety harnesses. To actively protect workers, a safety jacket is proposed. The jacket is equipped with an airbag cushion and an inertial measurement unit, located on the back, to sense the body motion hl. An algorithm deploys the airbag automatically for falls from heights greater than 1 m, leaving enough time for the airbag to be fully inflated. The proposed algorithm is tested and validated against experimental data collected during real falls performed by a professional stuntman, and during months of daily routine workers' life. Particular attention is devoted to the calibration of the tuning parameters, with the aim of the best trade-off between false positives' minimization, reactiveness, and robustness.